A Fuzzy Logic Control Technique for Blood Glucose Level in Diabetics

M.S. Ibbini, M.A. Masadeh, and M.M. Bani Amer (Jordan)

Keywords

Fuzzy Logic Control (FLC), PID, Continuous insulin infusion, Glucose mealdisturbance, White noise, minimal model of Bergman.

Abstract

A closed-loop system is proposed to control the plasma glucose levels of patients with diabetes mellitus type I. Biomedical systems are inhernetly complex, nonlinear, time varying, and may also involve time delays. In fact, conventional control techniques, often based on linearization, might prove inadequate and suffer from their dependency on mathematical models and experimental parameters. Model parameters values are often variable from patient to patient and exact parameters values are difficult to adopt. A fuzzy logic control scheme is proposed as a means of maintaining the normoglycemic average of 4.5 mmol/L as well as normal conditions, for free plasma insulin concentration, in spite of glucose like meal-disturbance and variations in model parameters. Computer simulations are used to demonstrate the efficiency of fuzzy logic controllers in handling disturbances and severe patient conditions. The results are compared to those resulted from well-known conventional controllers such as Proportional-Integral-Derivative (PID) and continuous insulin infusion control strategies.

Important Links:



Go Back